Does Hero Testing make us Testing Heros?
The plan has been approved and the team is fully committed, great stuff! Hang on a minute, what if… How about… hmm, I need another test! While I’m at it, it would be good to tidy up a few loose ends and address a few concerns. I’ll fetch my cape, it’s time for some Hero Testing; it can only make things better.
Jon first introduced me to the term ‘Hero Testing’ and to some of the challenges this ‘I can do it quick & cheap’ approach may bring. For example:
- Employ different process
- Compromise test repeatability (manual v automated test execution)
- Lead to conflicting priorities
- Undermine the value of ‘core’ testing already in plan
If this stuff is important, shouldn’t it be evaluated, prioritized, owned and managed by the team from *their* backlog? Maybe this is what a true Testing Hero would do. Shame, I thought I made the cape look good.
Tags: hero-testing, techniques
January 23rd, 2009 at 10:24
Good post Rich, get out the cape again…
As Rich says, we introduced (or recognized) the concept of ‘Hero Testing’ a few years ago as being the sort of testing that gets lots of project focus when its being done (wow – that tester is a real hero, what would we do without her?) but builds a legacy for the mere mortals who have to pick up the pieces later.
I do however feel that there is still a place for good, focused, possibly one-off manual testing, but this should be considered as part of the overall plan and be balanced by some sort of regression capability.
Manual testing is another tool in the risk-reduction kit bag.
The only sensible way to do this is to have it included in the team backlog, as Rich proposes.